Loading...

Table of Content

    16 July 2014, Volume 0 Issue 7
    A Structured System of Mining Chinese Product Reviews
    YANG Hui1, LIU Hong-yan2, HE Jun3
    2014, 0(7):  1-7+15.  doi:10.3969/j.issn.1006-2475.2014.07.001
    Asbtract ( 191 )  
    References | Related Articles | Metrics
    We designed and implemented a structured system of mining Chinese product reviews. According to the user’s specified product names, the system could grab reviews about the product on Internet, extract product features and automatically determine the polarity of context-dependent opinions and then return a structured result of the reviews as the triple (feature, opinion, polarity). The experiment proved the effectiveness of the algorithms. And it made an acceptable precision and recall.
    Combined Nested Partitions Method Based on Improved Genetic Algorithm
    ZONG De-cai1, WANG Kang-kang2, DING Yong3
    2014, 0(7):  8-15.  doi:10.3969/j.issn.1006-2475.2014.07.002
    Asbtract ( 266 )  
    References | Related Articles | Metrics
    This paper puts forward an improved nested partitions method using improved genetic algorithm to solve the small and medium scale traveling salesman problem. The algorithm adopts weighted sampling method to generate the initial most promising region, uses the global array to record the historical optimal solution of every region and designs the crossover operator and mutation operator of subregion. Then our algorithm uses an improved genetic algorithm to search the optimal solution of each subregion and surrounding region. In the search for the best solution of surrounding region, some excellent individual is improved by modified Lin-Kernighan algorithm. The simulation results for the 16 problems in TSPLIB show that the proposed improved nested partitions method using improved genetic algorithm can find solutions of high quality when applied to the traveling salesman problem.
    A Novel Validity Index for Fuzzy Clustering
    TANG Guan-bao
    2014, 0(7):  16-18+23.  doi:10.3969/j.issn.1006-2475.2014.07.003
    Asbtract ( 213 )  
    References | Related Articles | Metrics
    Combining compactness and separation information of fuzzy clustering, this paper proposes a novel validity index for clustering. This index can determine optimal partition and optimal number of clusters for fuzzy partitions obtained from the fuzzy C-means algorithm (FCM). Comparison experiment is done in one synthetic data set and four real datasets, and the results prove that this index has superior effectiveness.
    Search Engine of Massive Source Code Based on Semantic Network
    HU Xiang1, SHU Li-lian2
    2014, 0(7):  19-23.  doi:10.3969/j.issn.1006-2475.2014.07.004
    Asbtract ( 212 )  
    References | Related Articles | Metrics
    This paper proposes an analyzing and searching method which is capable of searching massive source codes whose function is close to the natural language input. Compared with traditional massive source code search engine based on keyword matching algorithm and source code location algorithm based on natural language, this approach solved the mismatch problem in keyword matching algorithm caused by abbreviation and similarity word in source code and the limitation to a specified project of later solution by innovative approach. Experiment is practiced to prove the reliability and usability of this engine.
    An Improved Text Clustering Algorithm Based on Latent Semantic Indexing
    HOU Ze-min, JU Xiao
    2014, 0(7):  24-27.  doi:10.3969/j.issn.1006-2475.2014.07.005
    Asbtract ( 211 )  
    References | Related Articles | Metrics
    This paper presents an improved text clustering algorithm based on latent semantic indexing. This algorithm introduces the theory of latent semantic index, improves the traditional SOM algorithm. By using the latent semantic indexing text feature vector representation theory, we mine the semantic structure relationships hidden among the words in text, thereby eliminating the correlation among words, to reduce the feature vector dimension. The limitations of the traditional SOM algorithm are improved to accurately give the number of clustering classes. Experimental results show that the clustering effect of this algorithm is better, and the clustering time is less.
    Similar Air Material Demand Forecasting Based on Case-based Reasoning
    DING Guang-wei, ZHOU Wei, CHENG Rui, ZHANG Bing
    2014, 0(7):  28-30+35.  doi:10.3969/j.issn.1006-2475.2014.07.006
    Asbtract ( 228 )  
    References | Related Articles | Metrics
    The naval aviation is developed very quickly, more new aircraft types appeare. According to the development of aviation material with the characteristics of homology and inheritance, by making a deep analysis of characteristics of aviation material and applying the theory of case-based reasoning, we construct a model of similar air material determination. And the scientificity of the method is verified by an example, providing a theoretical basis for aviation material support of the new aircraft types.
    Transfer Function Design Based on K-means Clustering
    CHEN Ji-wei, FAN Yin, XIANG Jun-li, LI Qian, ZHOU Rui, YANG Tong-man
    2014, 0(7):  31-35.  doi:10.3969/j.issn.1006-2475.2014.07.007
    Asbtract ( 213 )  
    References | Related Articles | Metrics
    In the direct volume rendering, transfer function defines the mapping rules between data property and the optical property, so it determines the effect of the image. The traditional multidimensional transfer functions are based on volume data and gradient data, this paper proposed a method based on K-means clustering. Furthermore, it realized weather data sorting and based on which it realized direct volume rendering based on GPU programming.
    Query Expansion Based on Interest Model
    TIAN Yong-chang1, LI Ying2
    2014, 0(7):  36-39.  doi:10.3969/j.issn.1006-2475.2014.07.008
    Asbtract ( 226 )  
    References | Related Articles | Metrics
    To solve the problem that words mismatch between the retrieval and the documents in the query information, this paper proposes a query expansion method based on interest model, and it is combined with conceptual vector space model, and can judge whether it is feasible to query expansion according to the threshold value. In this method, user’s interest is mined by Agent in real time, and the intellectual interests are expressed by the HNC (hierarchical network of concepts). As a result, it is very convenient to calculate the similarity between concepts. Experimental result shows that the performance has 29.1% improvement by using the query expansion compared to the non-query expansion.
    An Approach for Use Case Modeling Based on Goals Conversion
    LIN Yu-feng, CHEN Zhong-yu, LUO Zheng-ping, WU Xing-tong
    2014, 0(7):  40-44.  doi:10.3969/j.issn.1006-2475.2014.07.009
    Asbtract ( 257 )  
    References | Related Articles | Metrics
    In the approaches of goal-based requirements modeling, the over-dependence of natural languages has led to low efficiency and accuracy of modeling in the process from goals to requirements. This paper introduces XML and turns the goals into XML description to present the XML conversion algorithm, and clarify the concerns as well as their relationships, then to achieve the use case model. This modeling process avoids the ambiguity of natural languages.
    A Whole Domain Component Model System Based on Interface Specification of Beida Jadebird
    PAN Zhong-kui1, LIAN Shi-di2, ZHAO Qi1
    2014, 0(7):  45-48.  doi:10.3969/j.issn.1006-2475.2014.07.010
    Asbtract ( 162 )  
    References | Related Articles | Metrics
    When building domain code components, the existing component model is difficult to define and describe the family commonality and variability of the domain, resulting in component particle size is difficult to control. A whole domain component model system based on interface specification of Beida Jadebird in treeprograms is proposed. The system consists of the whole domain component modeland secondary processing tool which is used to process the corresponding components. By the whole domain component model, the family commonality and variability of the domain can be described in code. And, the secondary processing tool can be used to process the corresponding components, removing redundant sub-modules in the corresponding component. The system can effectively solve the problems of reusing and scale of the components which are frequently encountered in developing code components.
    Data Acquisition and Storage System Based on Online Compression
    FU Yuan-di, ZHANG Yan-yuan, LIN Yi
    2014, 0(7):  49-52+57.  doi:10.3969/j.issn.1006-2475.2014.07.011
    Asbtract ( 180 )  
    References | Related Articles | Metrics
    In high-speed data acquisition and storage system, storage rate is expected higher and higher. Compared with the high performance of computing capability and network bandwidth, storage system has become the bottleneck of the whole system. This paper implements a data storage system. To achieve high integrity, this system implements a set of simple API and allows developers to use the specified compression algorithm. Finally, by the test, the system improves the data rate significantly, as well as the utilization rate of storage space, under the condition of a proper compression ratio.
    Parallel Computing of Infrared Small Target Detection Based on MPI and OpenMP
    HE Wei-wei, WU Jing, ZENG Yao-yuan
    2014, 0(7):  53-57.  doi:10.3969/j.issn.1006-2475.2014.07.012
    Asbtract ( 161 )  
    References | Related Articles | Metrics
    To get effective surveillance of the whole movement of infrared small target, multiband simultaneous detection is necessary. But, multiband simultaneous detection leads to substantial growth in computing time, that cannot meet the requirement of real-time target detection in practical application. To solve this problem, this paper presents a parallel hierarchical approach based on MPI and OpenMP, which makes full use of the advantages of both message-passing model and shared storage model. It is based on multi-processor node cluster for testing. Experimental results show that, this parallel version of the program can achieve a speedup of 8.61 compared to the serial version in the condition of same detection probability. It improves the efficiency of target detection greatly.
    Multiple Instance Target Tracking Algorithm Based on Compressed Sense
    CHEN Xi, DI Lan
    2014, 0(7):  58-62+67.  doi:10.3969/j.issn.1006-2475.2014.07.013
    Asbtract ( 184 )  
    References | Related Articles | Metrics
    Aiming at the problems of unstable tracking, easy to drift, obscured loss, which are produced in real-time target tracking, we propose an improved tracking algorithm for multiple instances. In the compressed sensing and real-time tracking, by adding random measurement matrix to produce new features, multiple positive and negative instances are integrated. By combining with the boosting learning method to update the feature weights and improve the confidence map estimation, we solve the problems of target drift and loss. Experimental results show that the proposed algorithm achieves better robustness and stable real-time tracking when the target moves quickly, or in conditions that the textures and lightings change seriously, as well as it is partially covered.
     Image Matching Algorithm Based on SIFT
    ZHOU Li-fen
    2014, 0(7):  63-67.  doi:10.3969/j.issn.1006-2475.2014.07.014
    Asbtract ( 212 )  
    References | Related Articles | Metrics
    Key points are very few when extracted by the SIFT algorithm directly and mismatch points are much more during the image matching of the next step from blur image. We propose a matching algorithm based on SIFT feature. Firstly, we use the Laplacian operator to sharpen the blurred image in order to highlight the edge, and then extract the key points by the SIFT algorithm. Lastly, we use the bidirectional matching algorithm to match image. The experimental results show that the number of matched points by the improved algorithm has more points than the way of using SIFT algorithm directly and the results of matching are better.
    Quantum State Model of Fuzzy Morphological Structuring Element
    ZHANG Cheng-bin, WANG Kai-fu
    2014, 0(7):  68-70.  doi:10.3969/j.issn.1006-2475.2014.07.015
    Asbtract ( 254 )  
    References | Related Articles | Metrics
    Quantum computing as a new computational theory model is becoming the focus of information study. To realize the effective fusion of fuzzy morphology and quantum computing model, a fuzzy morphological structuring element based on quantum state model is proposed. Inspired by quantum signal processing theory, the structuring element is denoted by quantum bit system, a degree of membership function denotes a probability of quantum state. A fuzzy structuring element is collapsed as a clear structuring element through quantum measurement. Simulation analysis shows the feasibility of the quantum state model, and morphological filter can effectively remove noise and keep the detail information by using the collapsed structuring element.
    Large-scale Facial Feature Retrieval Based on Cloud Computing
    ZHANG Zhi-qiang
    2014, 0(7):  71-74+79.  doi:10.3969/j.issn.1006-2475.2014.07.016
    Asbtract ( 232 )  
    References | Related Articles | Metrics
    In the process of matching facial images, the traditional algorithm compares each face’s features one by one. With the increase of images, the comparison time prolongs which leads to the low match efficiency. A match algorithm of mass facial feature based on cloud is proposed. First, the weighting parameters of facial features are extracted by inter-cloud network models. Then, the transformation processing of the features’ parameter matrix is conducted. Last, this matrix is applied to matching facial features. The cloud platform based on this method combines the features of the improved ASM algorithm’s early-detection model which helps match. Experimental result shows that the suggested algorithm has the advantage of less time-consuming compared with the traditional one in the match of mass facial images.
    Gesture Tracking and Motion Recognition Algorithm Based on Vision Technology
    XIONG Jun-tao1, LIU Zi-jian1, SUN Bao-xia2, YU Shou-hua1, CHEN Jian-guo1
    2014, 0(7):  75-79.  doi:10.3969/j.issn.1006-2475.2014.07.017
    Asbtract ( 269 )  
    References | Related Articles | Metrics
    The gesture tracking and motion recognition of production line disciplined operational motion in industrial engineering was researched in this paper. Through the color analysis of hands skin, the YCbCr color model was used to recognize the hand area, the pretreatment so as threshold segmentation for video image was used in YCbCr color model to obtain the complete hand area; the Euclidean distance transform method was used to calculate the distance between the adjacent two hand trajectory points and speed of each frame image's hand movements, then the extended finite state machine model was used to realize the segmentation of hand movements, matching the segmented motions and established motion templates, and the Hausdorff distance matching method was used to determine the accuracy of the matching results, the recognition of hand gesture was realized. Experimental results show that the hand gesture recognition algorithm has certain anti-noise ability for background interference and camera vibrations, and has higher recognition accuracy, can satisfy the application requirements of real work environment.
    An Improved GASA Algorithm for Image Segmentation
    TAN Ding-ying, CHEN Ping-ping, LI Xue-zheng, TANG Rong
    2014, 0(7):  80-84+88.  doi:10.3969/j.issn.1006-2475.2014.07.018
    Asbtract ( 261 )  
    References | Related Articles | Metrics
    Image segmentation is the foundation of the image processing and analysis. The Otsu segmentation algorithm and genetic algorithm are analyzed in this paper, in order to improve the running performance of the algorithm, simulated annealing is introduced to put forward a kind of improved genetic simulated annealing algorithm (GASA). The whole running process of this algorithm was controlled by the temperature cooling schedule, with the improved Otsu method being used as the fitness function of the genetic algorithm. After several rounds of computing, an optimal threshold value was obtained for image segmentation. The experiments’ results showed that the image segmentation based on the GASA could be good at enhancing the comprehensive search ability of the algorithm, and avoiding the genetic algorithm’s falling into local optimization. Meantime, it would not only converge to the optimum segmentation threshold faster and more steadily, but also obtain higher segmentation quality.
    Image Inpainting Algorithm Based on Structured Dictionary
    LIU Chun-rong
    2014, 0(7):  85-88.  doi:10.3969/j.issn.1006-2475.2014.07.019
    Asbtract ( 229 )  
    References | Related Articles | Metrics
    With the improvement of sparse representation theory, it is widely used to make full use of sparse property of signals to repair the image. The traditional dictionary has the problem that it cannot make full use of the correlation between dictionary atoms and just a collection of unstructured atoms. Therefore we propose an image inpainting algorithm based on structured dictionary and illustrate the validity in image inpainting tasks. The algorithm can not only finish image inpainting task by training tighten dictionary, but also the dictionary has the merits of shift-invariance and scale flexibility and so on.
    Facial Expression Recognition Techniques Based on Bilinear Model
    XU Huan
    2014, 0(7):  89-93.  doi:10.3969/j.issn.1006-2475.2014.07.020
    Asbtract ( 214 )  
    References | Related Articles | Metrics
    Aiming at the problems existing in facial expression recognition currently, based on the data in the 3D expression database BU-3DFE, we study the point cloud alignment of 3D facial expression data, establish the bilinear models based on the alignment data, and improve the recognition algorithms based on bilinear model in order to form the new recognition and classification algorithms, to reduce the quantity of identity feature calculation in original algorithm, to minimize the influence of identity feature on the total expression recognition process, to improve the results of facial expression recognition, and to ultimately achieve the high robustness of 3D facial expression recognition.
    Plug-in Type Intelligent Anesthesia System
    WANG Can-can1, ZHAO Feng-yu1, WANG Yong2
    2014, 0(7):  94-97+103.  doi:10.3969/j.issn.1006-2475.2014.07.021
    Asbtract ( 209 )  
    References | Related Articles | Metrics
    It is meaningful to examine anesthetic program process and optimize anesthetic method in anesthesia system. In this paper, an intelligent system with expert knowledge base and optimal anesthetic case base is provided and plugged into anesthesia system. The anesthetic expert knowledge base is based on expert knowledge, experience and anesthetic rules which are mainly used for examining anesthetic plan given by anesthetists. The anesthetic program is further optimized based on successful anesthetic case base. The intelligent anesthesia system will be integrated into the existing surgery anesthesia management information system as a plug-in component.
    Remote Monitoring System for Electric Vehicle
    GONG Wen, ZHANG Hui, XIONG Zhi
    2014, 0(7):  98-103.  doi:10.3969/j.issn.1006-2475.2014.07.022
    Asbtract ( 290 )  
    References | Related Articles | Metrics
    A remote monitoring system for electric vehicles based on GPRS wireless network is designed and implemented, which can integrate the GPS data from the electric vehicle monitoring terminal, the battery and the motor data from the vehicle's CAN bus and send them to the monitoring center. The submitted data is analyzed in monitoring center by custom data transfer protocol and the remote real-time monitoring of vehicle running status is realized through post analysis and processing.
    A Mobile Phone Sign-in System Based on Android
    DI San-yun, SONG Li-hong, ZHANG Kai
    2014, 0(7):  104-108.  doi:10.3969/j.issn.1006-2475.2014.07.023
    Asbtract ( 249 )  
    References | Related Articles | Metrics
    The paper presents a mobile terminal application based on text recognition and embedded database technology. The application can conveniently record, manage and upload attendance report, generate the attendance results. The application improves the quality of teaching with low cost and high efficiency.
    Modeling of Computer Network Cascading Failure Based on Load Capacity
    WANG Hong-wei, ZHAO Hui-na
    2014, 0(7):  109-112+116.  doi:10.3969/j.issn.1006-2475.2014.07.024
    Asbtract ( 242 )  
    References | Related Articles | Metrics
    Based on computer system initialized load capability, routing control strategy and node forwarding rate, in view of the cascading failure influence in computer system network service performance, we construct a kind of computer network cascading failure model considering the service performance. From multiple levels of influence parameters, the model effectively measures the influence factors variable values of cascading failure for the computer network service performance. Through comprehensive analysis, the model could offer certain practice guidance for preventing and controlling the network cascading failure.
    An Improved Dijkstra Algorithm Based on Direction Priority and Bidirectional Search
    TANG Cai-hong
    2014, 0(7):  113-116.  doi:10.3969/j.issn.1006-2475.2014.07.025
    Asbtract ( 231 )  
    References | Related Articles | Metrics
    In the process of searching the shortest path, the traditional Dijkstra algorithm needs to traverse all vertices one by one in the network graph. The amount of calculation and storage space is large, and the search efficiency is very low. Therefore, based on the spatial characteristics of traffic network and the shortage of the traditional algorithm, storage structure is improved, and the combination of direction priority and bidirectional search is adopted. The purpose is to reduce the storage space and narrow search scope. As a result, search speed is accelerated, and search efficiency is raised. The experimental data shows that: compared with the traditional algorithm, the improved algorithm can search more effectively the shortest path in a transportation network, and has better practical value.
    Upper Bounds of Cluster-tree Wireless Sensor Network QoS Based on Strict Priority Scheduling Strategy
    CHEN Yue-yang, GU Ping, ZHANG Chao
    2014, 0(7):  117-120.  doi:10.3969/j.issn.1006-2475.2014.07.026
    Asbtract ( 225 )  
    References | Related Articles | Metrics
    In order to improve QoS (quality of service) of WSN (wireless sensor network) under the worst-case condition, which is particularly relevant for those time-sensitive WSN application, this paper applies the SP (strict priority) scheduling strategy to the cluster-tree WSN and analyzes how the strategy will impact the service guaranteed to data flows. The closed-recurrent expressions of calculating the worst-case end-to-end delay and queue length are derived based on DNS (deterministic network calculus). Finally, we illustrate that the scheduling strategy can improve the maximum end-to-end delay, while increase buffer requirements of sensor nodes, with numerical analysis.
    User Management Model on Flat Ethernet
    LIU Zhi-gang, ZHU Ling-hong
    2014, 0(7):  121-124.  doi:10.3969/j.issn.1006-2475.2014.07.027
    Asbtract ( 232 )  
    References | Related Articles | Metrics
    Flat Ethernet is a new network model based on multiservice router. In this model, the mechanisms of PUPV and IPoE are considered to be appropriate ways to solve the problems of user management on traditional Ethernet, such as user access security, user management and user maintains. This model is implemented on a real campus network, and improves user management and maintenance efficiency in fact.
    Petri Nets Node Refinement and Its Application
    QI Fang-min
    2014, 0(7):  125-128+132.  doi:10.3969/j.issn.1006-2475.2014.07.028
    Asbtract ( 202 )  
    References | Related Articles | Metrics
    A solution of Petri nets node explosion problem is given based on the concept of node refinement on Petri nets, and from this definition and the bound condition of place and transition, we analyze the feasibility and methods of node refinement on Petri net, which is helpful on modeling and verification with Petri net later.
    Regular Expression Grouping Algorithm Based on Expansion Coefficient
    WANG Mei-yang1, TANG Xue-wen2, YANG Zheng-yi3
    2014, 0(7):  129-132.  doi:10.3969/j.issn.1006-2475.2014.07.029
    Asbtract ( 269 )  
    References | Related Articles | Metrics
    Matching speed of det­erministic finite automata (DFA) is far faster than that of the non-deterministic finite state autom­ata (NFA). However, when a large number of regular expressions transform in­to a DFA, the D­FA may consume huge storage space because of state explosion. Firstly, to describe characteristics of the regular expressions, expansion coefficient was defined. Based on the concept of expansion coefficient, a m­ore efficient grou­p­­­ing algorithm which named IGA algorithm was prop­o­sed. IGA algorithm i­solated those regular expressions which easily lead to state explosion, thus saved st­orage space. Experimental results show that, compared with the original algorithm, when the number of groups is same to the original one, IGA algorithm could reduce 25 percents of the number of states averagely.
    A Digital Watermarking Algorithm of Color Images Based on Surface Fitting
    LU Jian, TU Guang-yi, WU Hai-yan
    2014, 0(7):  133-136.  doi:10.3969/j.issn.1006-2475.2014.07.030
    Asbtract ( 179 )  
    References | Related Articles | Metrics
    A digital watermarking algorithm for color images is discussed. The proposed method applies a self-adaptive segmentation technique to divide an image into three classes with different sub-block sizes, such as 16×16, 8×8 and 4×4, and chooses those blocks with size of 4×4 to embed the watermarks. In order to get robust performance, the method scrambles the 4×4 block by a pseudo random scheme, then embeds the watermarks to the scrambled blocks. To reduce the error that caused by surface fitting, we use error compensation technique to improve the quality of the watermarked image. The results show that the algorithm is robust against various typical attacks, such as compression, filtering, noising and shearing.
    Self-adaptive Image Encryption Algorithm Based on Hyper-chaotic System
    LI Yi-lin
    2014, 0(7):  137-141+145.  doi:10.3969/j.issn.1006-2475.2014.07.031
    Asbtract ( 238 )  
    References | Related Articles | Metrics
    This paper presents a self-adaptive image encryption algorithm based on hyper-chaotic system. Firstly, the hyper-chaotic sequences which are pretreated are transformed according to the plaintext pixel information that result in both key sensitivity and plaintext sensitivity; then, the last encrypted pixel is used to produce intermediate key through nonlinear operation to encrypt the next pixel adaptively, until all pixels are encrypted completely; finally, the simulation analyses are performed using statistical property, anti-differential attacks, key sensitivity and key space and algorithm efficiency. Experimental results show that the algorithm is secure and efficient.
    Parallel Password Recovery Framework Using MPI
    JIANG Lu-jin
    2014, 0(7):  142-145.  doi:10.3969/j.issn.1006-2475.2014.07.032
    Asbtract ( 240 )  
    References | Related Articles | Metrics
    Using passwords to protect data is the most widely deployed method for securities. When people need to recover encrypted data for easily guessable passwords, it can require millions of Hash function and string comparison operations. These operations can be computationally expensive but are easily parallelizable because each password can be tested independently. Therefore, using high performance computing (HPC) can greatly reduce the time required to perform password recovery. This paper designs P2RF: A parallel password recovery framework using MPI. P2RF can distribute decryption task into different computing nodes. Then, the computing node distributes tasks into different PEs. The experimental results show that: P2RF is linearly scalable.
    Application of Wavelet Decomposition in Steel Strip Defect Detection
    XU Shuai-hua, GUAN Sheng-qi, CHEN Long-long, LI Ting
    2014, 0(7):  146-149.  doi:10.3969/j.issn.1006-2475.2014.07.033
    Asbtract ( 210 )  
    References | Related Articles | Metrics
    The influence of illumination reduces the quality of strip image, a new kind of steel strip detection method is put forward through analysis of strip defect characteristics. First of all, the logarithm of strip image gray value is decomposed into a series of sub-graph through the wavelet transform. Secondly, every sub-graph of wavelet decomposition is treated by homomorphism, and then the center-surround difference operation is used to construct difference sub-map. On this basis, difference sub-maps are selected for image fusion and exponential operation is used to get the high contrast defect image. Finally, steel strip defect is detected through the segmentation method of maximum between-cluster variance. The result of experiments shows that this method can enhance the defect image contrast with clear image details, inhibit the effect of noise, and effectively realize the rapid detection of strip defect image.
    Realtime Simulation of ROV Movement Based on UDK
    LI Shen, LIAO Hua-li
    2014, 0(7):  150-153+158.  doi:10.3969/j.issn.1006-2475.2014.07.034
    Asbtract ( 323 )  
    References | Related Articles | Metrics
    In order to let the observers “be personally” on the working environment of ROV, this paper develops a three-dimensional virtual reality platform that vividly presents the working situations of the vehicle by utilizing the feedbacks of it. With the development tools of 3DS Max, UDK and VS2010, UnrealScript is used to reprogram the TCP/IP class to get the pose and speed level information for a further controlling of the 3D models of the vehicle structure, steering engines and propellers to show the work of ROV at the moment.
    Simulated Analysis on Scheme of Enterprise Network Expansion and Upgrading Based on OPNET
    LIU Qing, LIU Kai-fen
    2014, 0(7):  154-158.  doi:10.3969/j.issn.1006-2475.2014.07.035
    Asbtract ( 206 )  
    References | Related Articles | Metrics
    This paper puts forward the theory that OPNET can be used to make simulated analysis on the scheme of network expansion and upgrading aiming at the unpredictability of the effect of enterprise network when it is expanded and upgraded, and then estimates the feasibility of the scheme of network expansion and upgrading by analyzing the change of the key performance parameters before and after expansion and upgrade of the network. Practice shows that using OPNET to make a simulated analysis on the scheme of network expansion and upgrading can not only lower the risk of network expansion and upgrading greatly, but also verify the feasibility of the scheme of network expansion and upgrading efficiently.